HANDY WAVEGUIDE TXRF SPECTROMETER FOR NANOGRAM SENSITIVITY

Size: px
Start display at page:

Download "HANDY WAVEGUIDE TXRF SPECTROMETER FOR NANOGRAM SENSITIVITY"

Transcription

1 213 HANDY WAVEGUIDE TXRF SPECTROMETER FOR NANOGRAM SENSITIVITY Shinsuke Kunimura and Jun Kawai Department of Materials Science and Engineering, Kyoto University, Sakyo-ku, Kyoto , Japan ABSTRACT A specimen containing nanograms of sulfur, calcium, and 3d transition metal elements was measured by incident X-ray beams of various sizes restricted by a waveguide placed in a portable TXRF spectrometer. The signal to background ratios of spectra decreased with an increase in incident X-ray beam size. The portable spectrometer was also applied to rainwater and a specimen containing antimony and rare earth elements. Nanograms of elements in these specimens were detected by K-line or L-line excitation. INTRODUCTION X-rays were totally reflected on a flat surface, and this phenomenon was first reported in 1923 [1]. Formulas for X-ray reflectivity were first presented in 1954 [2]. In 1971, XRF analysis using X-ray total reflection was first proposed [3]. The details of TXRF analysis were described in the late 1990s and 2000s [4,5]. Matrix effect was negligible in the TXRF analysis, and the fluorescent X-ray intensities were linear with the concentration of elements [6]. Using a low-pass filter could decrease background intensities, and a TXRF spectrometer with a double quartz reflector as a low-pass filter achieved several pg detection limits [7]. A combination of a monochromator and a high-power X-ray source such as a rotating X-ray tube [8] and synchrotron radiation [9] was effective for increasing the signal to background ratios. The detection limit of 13 fg (10-15 g) for a 3d transition metal element was achieved by using monochromatic synchrotron radiation [10]. Monochromatic synchrotron radiation was also effective for detecting trace quantities of low Z elements such as sodium and magnesium, and several tens of fg detection limits were obtained by using a detector with an ultra thin window [11]. As well as a high-power X-ray source described above, a low-power X-ray tube (maximum load of 40 or 50 W) was also used to perform highly sensitive analysis by a combination of a monochromator and a portable TXRF spectrometer [12,13]. The detection limits by these two spectrometers were a few pg or a few tens of pg. A portable spectrometer with a planer waveguide and a 50 W X-ray tube was reported [14]. This combination in the spectrometer achieved a few tens of pg detection limit. Usefulness of a waveguide consisting of two parallel plane reflectors for the TXRF analysis was also reported [15]. An X-ray optic consisting of two plane reflectors as well as a waveguide is useful for simple optical arrangement in TXRF analysis [16]. A portable TXRF spectrometer was designed, constructed, and reported recently [17]. A monochromator was not used, and the excitation source was the continuum X-rays from a 1.5 W X-ray tube. A waveguide was used as an X-ray optic, and the length of it was about 1 cm. The

2 This document was presented at the Denver X-ray Conference (DXC) on Applications of X-ray Analysis. Sponsored by the International Centre for Diffraction Data (ICDD). This document is provided by ICDD in cooperation with the authors and presenters of the DXC for the express purpose of educating the scientific community. All copyrights for the document are retained by ICDD. Usage is restricted for the purposes of education and scientific research. DXC Website ICDD Website -

3 214 distance between the X-ray tube and a specimen was about 4 cm because of the short length of the waveguide. Attenuation of the X-rays as the excitation source was decreased by this positioning, and 1 ng detection limit was achieved. The portable spectrometer has been applied to leaching test solutions of soils [18] and commercial bottled drinking water [19], and nanograms of elements in these specimens were detected. In the present paper, a specimen containing nanograms of sulfur, calcium, and 3d transition metals was measured by the incident X-ray beams of three different sizes which were restricted by a waveguide, and the signal to background ratios of TXRF spectra were compared. Applications to a specimen containing rare earth elements and rainwater are also presented. SPECTROMETER The details of the portable spectrometer are reported in a previous paper [17], but are summarized as follows: (1) the continuum X-rays from a tungsten target X-ray tube, which a voltage of 9.5 kv was applied, was used as the excitation source, (2) the incident X-ray beam was collimated by a waveguide, (3) fluorescent X-ray intensities were measured by a Si PIN detector containing a preamplifier and a digital signal processor in the detector enclosure, (4) the number of channels in this multi-channel analyzer was 512 and 1 channel corresponded to 38 ev of X-ray energy, and (5) a quartz optical flat (30 mm in diameter and 10 mm in thickness) was used as a sample holder. Figure 1 shows a photograph of the portable spectrometer. Figure 1. A portable TXRF spectrometer. SAMPLE PREPARATION AND MEASUREMENTS A mixed standard solution containing 0.1 ppm of Ca, Sc, Ti, V, Cr, Mn, and Fe and 0.3 ppm of Ni (Sample 1), a mixed standard solution containing 0.5 ppm of Sb, Sc, and La, 1 ppm of Nd, 2 ppm of Eu, 5 ppm of Tb, and 10 ppm of Er and Yb (Sample 2), and rainwater (Sample 3) were measured. Samples 1 and 2 were prepared from 1,000 ppm standard solutions of these elements (scandium and lanthanum standard solutions contained 100 ppm of scandium and 5 ppm of lanthanum). Rain drops were collected in a polypropylene bottle, and the elements in the rainwater were quantified by an internal standard element. Neodymium was used as the internal standard element, because it is a rare element and would not be present in the rainwater. The rainwater and a 100 ppm Nd standard solution were mixed in a 9:1 ratio. As given in Table I, several tens of µl portions of each sample were pipetted on optical flats and then dried. Table I also gives the diameters of each sample dry residue and the amount of elements in each dry residue.

4 215 Sample 1 was measured by the incident X-ray beam of sizes restricted to 15, 50, and 100 µm in height and 10 mm in width by the waveguide. Sample 2 and 3 were measured by the incident X- ray beam of sizes 50 µm in height and 10 mm in width. All measurements were performed in air for 600 s at the glancing angle of The critical angle of total reflection for 9.5 kev X-rays on the optical flat was All areas of the dry residues were illuminated by the incident X-ray beams. Table I. Sample volume, diameter of each dry residue, and the amount of elements in each dry residue. Sample 1 Sample 2 Sample 3 Sample volume 40 µl 20 µl 20 µl Dry residue diameter ~1 mm ~4 mm ~4 mm Amount of each element 4 ng of Ca-Fe 10 ng of Sb, Sc, La 200 ng of Nd 12 ng of Ni 20 ng of Nd, 40 ng of Eu 100 ng of Tb, 200 ng of Eu, Yb RESULTS AND DISCUSSION Figure 2 shows representative TXRF spectra of Sample 1. As shown in Figure 2, nanograms of Ca, Sc, Ti, V, Cr, Mn, Fe, and Ni in the dry residue were detected, as well as silicon, sulfur, and argon. Spectral background and the net intensities of the fluorescent X-ray peaks increased with an increase in incident X- ray beam size. Spectral background and the silicon Kα peak appeared because penetration depth into the optical flat of the incident X-ray beam was in nanometer scale even when a glancing angle was smaller than the critical angle. The background intensity and the net intensity of the silicon Kα peak increased with an increase in the area of the optical flat illuminated by the incident X-ray beam. The argon Kα peak appeared because air contained 0.93 vol% argon. The sulfur Kα peak appeared because the titanium and vanadium standard solutions contained 1.25 and 0.45 mol/l H 2 SO 4, respectively. The dry residue contained 218 ng of sulfur. Although the dry residue was fully Figure 2. Representative TXRF spectra of Sample 1 measured by the incident X-ray beam of sizes restricted to (a) 15, (b) 50, and (c) 100 µm in height. illuminated by the incident X-ray beams of these three sizes, the net intensities of fluorescent X- ray peaks of the elements in the dry residue increased with an increase in the incident X-ray beam size. There was a possibility that the intensity of the incident X-ray beam per unit area of the optical flat increased. In addition, the signal to background ratios decreased with an increase in the incident X-ray beam size (Figure 3). Representative TXRF spectra of (a) Sample 2 and (b) a blank optical flat are shown in Figure 4. Nanograms of Sb, La, Nd, Eu, Tb, and Er in the dry residue were detected by L-line excitation.

5 216 Ytterbium was not detected because the amount of ytterbium would be lower than the detection limit. Figure 3. Results of the ratios between the net intensities of Kα peaks of elements and background intensities under their peaks measured by the incident X-ray beam of sizes restricted to (a) 15, (b) 50, and (c) 100 µm in height. Results of the relative sensitivities for elements measured by the portable spectrometer are shown in Figure 5. The relative sensitivities for S, Ca, Sc, Ti, V, Cr, Mn, Fe, and Ni were determined by the spectrum in Figure 2(b), and those for Sb, La, Nd, Eu, Tb, Er, and Yb were determined by the spectrum in Figure 4(a). The relative sensitivity for scandium was defined as unity. As shown in Figure 5, the relative sensitivity for calcium was at a maximum in K-line excitation, and lanthanum was at a maximum in L-line excitation. Representative TXRF spectrum of Sample 3 is shown in Figure 6. Sulfur and calcium were detected. The quantified concentrations of sulfur and calcium in the rainwater were 0.8 and 0.2 ppm, respectively, which were equivalent to nanograms of sulfur and calcium in the dry residue. CONCLUSION Figure 4. Representative TXRF spectra of (a) Sample 2 and (b) the blank optical flat. A specimen was measured by incident X-ray beams of three different sizes restricted by a waveguide. The signal to background ratios of TXRF spectra increased with a decrease in the incident X-ray beam size, but fluorescent X-ray intensities decreased. A specimen containing

6 217 rare earth elements was measured. Although the relative sensitivities for Sb, La, Nd, Eu, Tb, and Er were 1 2 orders of magnitude lower than that for calcium, nanograms of these six elements were detected by L-line excitation. Figure 5. Results of the relative sensitivities for elements measured by the portable spectrometer. Figure 6. A representative TXRF spectrum of Sample 3. ACKNOWLEDGMENT The present research was financially supported by the Asahi Glass Foundation. REFERENCES [1] Compton, A. H. Philos. Mag. 1923, 45, [2] Parratt, L. G. Phys. Rev. 1954, 95, [3] Yoneda, Y.; Horiuchi, T. Rev. Sci. Instrum. 1971, 42, [4] Klockenkämper, R. Total Reflection X-ray Fluorescence Analysis; Wiley: New York, [5] Wobrauschek, P. X-ray Spectrom. 2007, 36, [6] Wobrauschek, P.; Aiginger, H. Anal. Chem. 1975, 47, [7] Knoth, J.; Schwenke, H. Fresenius J. Anal. Chem. 1980, 301, 7 9. [8] Iida, A.; Gohshi, Y. Jpn. J. Appl. Phys. 1984, 23, [9] Iida, A; Yoshinaga, A.; Sakurai, K.; Gohshi, Y. Anal. Chem. 1986, 58, [10] Wobrauschek, P.; Görgl, R.; Kregsamer, P.; Streli, Ch.; Pahlke, S.; Fabry, L.; Haller, M.; Knöchel, A.; Radtke, M. Spectrochim. Acta, Part B 1997, 52, [11] Streli, Ch.; Pepponi, G.; Wobrauschek, P.; Zöger, N.; Pianetta, P.; Baur, K.; Pahlke, S.; Fabry, L.; Mantler, C.; Kanngießer, B.; Malzer, W. Spectrochim. Acta, Part B 2003, 58, [12] Waldschläger, U. Adv. X-ray Anal. 2000, 43, [13] Yamada, T.; Matsuo, M.; Kawahara, N.; Shimizu, Y.; Mantler, M. Monitoring of Sn and Fe impurity-densities in glass surfaces with a bench-top TXRF spectrometer. 12th Conference on Total Reflection X-Ray Fluorescence Analysis and Related Methods, Trento, Italy, June 2007.

7 218 [14] Jiménez, R. E. A. Bench top X-ray fluorescence spectrometers based on orthogonal and total reflection geometry for excitation. European Conference on X-Ray Spectrometry, Alghero, Italy, June [15] Egorov, V. K.; Egorov, E. V. Spectrochim. Acta, Part B 2004, 59, [16] Sánchez, H. J. Nucl. Instrum. Methods Phys. Res., Sect. B 2002, 194, [17] Kunimura, S.; Kawai, J. Anal. Chem. 2007, 79, [18] Kunimura, S.; Kawai, J.; Marumo, K. Adv. X-ray Chem. Anal., Jpn. 2007, 38, [19] Kunimura, S.; Kawai, J. Anal. Sci. 2007, 23,